Uranium and plutonium equilibriums in aqueous solutions to 200

Robert J. Lemire* and Peter R, Tremaine. Research Chemistry Branch ... used in a modified Criss-Cobble extrapolation scheme (13, 14), together with av...
0 downloads 0 Views 1MB Size
J. Chem. Eng. Data 1980, 25, 361-370

361

Uranium and Plutonium Equilibria in Aqueous Solutions to 200 "C Robert J. Lemire' and Peter R. Tremaine Research Chemistry Branch, Whiteshell Nuclear Research Establishment, Atomic Energy of Canada Research Company, Pinawa, Manitoba, Canada ROE 7L0

Standard molal Glbbs energy of formation and entropy data for aqueous uranium and plutonium ions in ail oxidation states and thelr complexes with OH-, Ci-, F-, Cot-, P043-,and SO4*- are critlcally reviewed. Selected values are used In a modified Crlss-Cobble entropy extrapolation to derive self-consistent analytical expressions for the temperature dependence of the standard molal Glbbs energles of formation of 74 aqueous species and, hence, the Glbbs energy changes and equlllbrlum constants of their complexation and solubility reactions, from 25 to 200 OC. Introduction The thermodynamic properties of dissolved uranium species and uranium minerals near room temperature have been extensively evaluated by geochemists interested in the formation of uranium ore deposits ( 7-7), most recently by Langmuir (8). Some authors have considered hydrothermal ore body formation in terms of a few selected uranium solution species ( 7, 5- 7 ) , but overall there has been only limited evaluation of the thermodynamic properties of uranium species at temperatves above 25 OC. The need for reliablle data at elevated temperatures has become much more crucial in recent years as attempts are now being made to ana1yz.e two new geological situations. These are the transfer of actinides by ground water from, flrst, the Oklo natural reactor in Gabon (9) and, second, from nuclear fuel or encapsulated high-level wastes in deep underground permanent disposal vaults (70). In both contexts, there is a need to understand the behavior of aqueous complexes of both uranium and higher actinides, in particular plutonium, at elevated temperatures. Rai and Swne ( 7 7) and Cleveland ( 72) have reported some information concerning the behavior of plutonium species at 25 OC, but they did not extend their work to higher temperatures. In the present paper, we present a critical evaluation of the thermodynamic data at 25 OC available in the literature for uranium and plutonium oxide, hydroxide, phosphate, sulfate, carbonate, fluoride, and chloride equilibria. The results are then used in a modified Criss-Cobble extrapolation scheme ( 73, 74), together with available high-temperature experimental data to estimate thermodynamic constants for these systems at temperatures up to 200 "C. Thermochemical Data Assessment Water and Potentia/ Ligands. Gibbs free energies of formation and entropies for all the species not containing uranium or plutonium were taken from the CODATA (75) or NBS ( 76) tables except as noted below. Heat capacity functions for gases and for those solid phases not containing uranium or plutonium are from Kelley's cornpilation (77). Values for the ionic dissociation product of water K ,, and for the apparent molar free energy of water under its own vapor pressure were taken from Olofsson and Hepler ( IS)and from Helgeson and Kirkham ( 79), respectively. Temperature-dependent constants for the protonation reactions of HP0-; in water are from Mesmer and b e s (ZO), and for SO;-, F-, and COC- from Marshall and Jones (27), 0021-9568/80/1725-0361801 .OO/O

Ellis (22), and Helgeson (23),respectively. Henry's law constants for C02 are also from Helgeson's tabulation ( 2 3 ) . Uranium. The thermochemical data base for the uranium system is detailed in Tables Iand 11. All of the data were critically evaluated. In cases where two values appeared equally plausible, we chose values consistent with Langmuir's comprehensive compilation (8). The source values for species considered here but not considered by Langmuir and for cases in which our choice of values for a species differed from Langmuir's are discussed below. The free energy and entropy of U02(s) were taken from Langmuir (8) as -1031.8 kJ mol-' and 77.03 J mol-' K-'. The values for the free energy and entropy of 7-U03and U308were taken from the same source. The enthalpy of formation of u4og as recalculated by Fuger (24)from the enthalpy of solution data of Fitzgibbon et al. (25) is -4513.7 kJ mol. The entropy of u4og is 334.2 J mol-' K-' as determined by using the heat capacity data of Grranvold et al. (26). When combined with CODATA values for So(U) and So(02)a value of AG,0(U40e)= -4278.4 kJ mol-' is obtained (marginally lower than the value given by Langmuir (8)). However, u4og undergoes a X transition near 75 OC with an associated transition enthalpy of 2.8 kJ mol-' and transition entropy of 8.72 J mol-' K-' (26). As we are primarily interested in the behavior of the uranium species at and above 100 OC, the enthalpy and entropy change for the transition were added to the 25 OC AHf' and So values to calculate the hypothetical AG,' = -4278 kJ mol-' and So = 342.9 J mol-' K-' for the hightemperatureform of UO , , at 25 OC. Cordfunke (27) has reported values for the enthalpies of solution of U03, U02(OH),, and U03QH20in 6 M "OB. The value for AH? of y-U03 is taken from Langmuir as -1223.8 kJ mol-' and the partial molar enthalpy of transfer of water to 6 M HN03(aq)is taken as -0.5, kJ mol-' (28, 29). Thus AHf0(UO3.2H20)is -1826.7 kJ mol-'. Langmuir has discussed the computation of a solubili product for UO,jOH), from the solubilii data of Gayer and Leider (30). Naumov et al. (6,37)have noted that the solid phase in Gayer and Leider's experiments was probably U03. 2H20, not U02(OH),, as the dihydrate is the stable form at 25 OC (32). Thus we have assumed that the solubility product (log K,' = 5.6) actualty refers to W3QH2O 2H+ F? U02* 3H20 and calculated AG,0(U03-2H20)= -1632.2 kJ mol-' accordingly. We have also assumed that AGRo = 0 for U03QH20 F! U02(OH)2 H20 at 60 OC (6)and used appropriate heat capacities and the Criss-Cobble principle (see below) to calculate S o and AG? for P-U03.H20. Values for the Gibbs energies and entropies for the simple U02+,and UOt+ ions were those of Fuger and Oetting U3+, (29)(as used by Langmuir). For the uranium(1V) aqueous hydrolysis species, the AGfo values given by Langmuir were adopted. Langmuir (8) has discussed the difficulties of estimating entropies of actinide solution species. The several empirical for species of the approaches (8, 33-35) for estimating type M(OH),(4-X'+yield very different results. For example, the treatment of Cobble ( 3 4 )leads to so(U(OH),-) = 318 J mol-' K-', and a treatment based on modification of b e s ' formula for the first hydrolysis constant (35) gives 485 J mol-' K-'. Langmuir estimates a value of 71 J mot-' K-' by assuming similar values for uranium complexes of the same charge, regardless of the oxidation state of the uranium. This last assumption is entirely arbitrary but is probably reasonable for

+

+

+

u+,

so

so

0 1980 American Chemical Society

362

Journal of Chemical and Engineering Data, Vol. 25, No. 4, 1980

complexes involving anions of a similar entropy per unit charge. Furthermore, the method is based on the behavior of uranium species rather than that of simple inorganic anions or of complexes of other metals. We, therefore, have adopted Langmuir's species. As the series of values for So for the U(OH)x(CX)+ hydrolysis products U(OH)X(kxwwas originally used by Baes and Mesmer (35)to circumvent considering illdefined species such as Ue(OH)159+,this last species was not considered in our analysis. The AGfo and So values for UO,(OH)+, (U02)2(0H)22',and (UO,),(OH),+ are those used by Langmuir (4, based on the work of Baes and Meyer (36). Sutton (37)has shown that an anionic hydrolysis species is formed in basic uranium(V1) solutions and has determined an equilibrium constant (log K = -30.7) for the reaction 3UOZ2+ 7H20 a (UO,),(OH,)7H+ at an unspecified but low ionic strength, from potentiometric measurements. This leads to AGIo((UO,),(OH),-) = -4343 kJ mol-'. Tsymbal's (38)much larger formation constant for (uo2)3(0H)7(log K = -24.0 at I = 0.1) is not consistent with solubility data for either ,8-UOz(0H), or U03QH20(6). Baes and Mesmer (35) have reported an empirical correlation of entropies of formation for polynuclear hydrolysis species. Using, in this formula, the value of the ratio of the charge to interatomic distance which it was correctly gives the experimental ASRo for (UO,),(OH),+, calculated that ASRo((U0,)3(OH)7-)= 101 J md-' K-' and hence S0((u0&(0H)7-) = 299 J mol-' K-l. In interpreting the solubilii of UO,(OH),(s) at high temperatures, Nikitin et al. (6) suggest that UO,(OH)&aq) is an important species. We accept Nikitin's value of 35 kJ mol-' for the free energy of solution of UO,(OH), (assumed to be @-U03.HpO)and l6., kJ mol-' for the enthalpy of solution, both at 25 OC. These lead to AGi0(UO,(OH),(aq)) = -1359 kJ mol-' and S'(UO,(OH),(aq)) = 59 J mol-' K-'. For UF,(s), more recent values than those used by Langmuir are available. Rand and Kubaschewski ( 3 9 ) estimated AH,O(UF,(s)) = -1882., kJ mol-'. A later value, due to Khanaev (40)has not been used as it depends on a value for AHfO(UCI,), a quantity which itself is the subject of debate (41, 42). So(UF4(s))is taken as 151.7 J mol-' K-' from the work of Burns et al. (43),giving AGfO(UF,(s)) = -1792 kJ mol-'. The AGIO of UF4-2.5H20was taken from Langmuir (8).The enthalpy of hydration of UF,(s) to UF4-2.5H,0 was found to be -34 kJ mol-' by Popov et al. (44),who used a welbracterized UF, sample. This gives AHI0(UF4.2.5H20) = -263., kJ mol-' and hence So = 289 J mol-' K-'. The Gibbs energies and entropies of the U4+and U O l + fluoride complexes at 25 OC are taken from Langmuir. Langmuir (8) has taken the values of Day et al. (45) (log 0'' = 2.52, log p2* = 3.87) for the complexation reactions U4+ nHS04- e U(S04),(C2"~ nH+ (n= 1, 2) at I = 2. The values for the complexation constants with sulfate ions (rather than with hydrogen sulfate ions) at I = 0 were found by applying the same corrections used by Wagman et al. (46)for the corresponding thorium complexes. This gave log p1 = 5.47, log p2 = 9.72, and AGio(USO:') = -1307 kJ mol-', AGIo(U(SO,),(aq)) = -2076 kJ mol-'. To obtain 3" for the thorium complexes, Wagman ( 4 6 ) used -122 J mol-' K-' for the entropy of deprotonation of the monohydrogen sulfate ion at I = 2. By use of this value, the CODATA value of So(SO:-) and ASRO = 3 J mol-' K-' for (U4+ HS04- e US04'+ -t H)' ( 4 5 ) gives S0(US042') = -272 J mol-' K-'. Similarly, Day's value (45), ASRO = 42 J mol-' K-' for the reaction of HS04-with USO': gives So(U(SO,),(aq)) = -1 10 J mol-' K-l. For UO,(SO,)(aq), Langmuir uses AGfo and So based on the solubility study of Leitzke and Stoughton (47). A recent potentiometric study by Nikolaeva (48, 491, done at lower ionic strength and over the temperature range 25-150 OC, is in basic agreement with the solubility study to about 100 OC. Above this temperature, Nikolaeva finds a markedly lower complexation constant for the

+

+

+

+

+

U O ~ ' l S 0 ~system, possibly because of inadequate consid+: hydrolysis equilibria in the earlier study (49). eration of the UO We have taken log p1= 2.93 for the complexation reaction from the work of Nikolaeva and obtain PGfO(UO2SO4(aq))= -1714 kJ mol-'. A least-squares fit of Nikolaeva's data yielded the 137Tln ( T l function RTIn p1 = 16350 - 6.8(T- 298.15) 298.15). This corresponds to a value for AS,' of 130.2 J mol-' K-' and, hence, So(U0,S04(aq)) = 52 J mol-' K-'. Only the monosulfate complex of uranyl ion was considered in this analysis, since further association to form anionic complexes is expected to be small ( 4 9 )at low sulfate ion concentrations (